A novel approach for one-step defect detection and depth estimation using sequenced thermal signal encoding
Source abstract
Abstract
Pulsed thermography is a technique of significant interest in non-destructive testing, particularly in defect detection and depth characterisation of composite materials. This study presents an innovative methodology for simultaneously detecting defects and estimating depth using a combination of sequenced thermal signal encoding and a two-dimensional convolution neural network (CNN) model. We compare the results of the proposed method with those obtained from the feed-forward neural network (FFNN), a one-dimensional CNN, and a long short-term memory recurrent neural network (LSTM-RNN). The findings demonstrate that the proposed approach exhibits superior accuracy and robustness compared to the benchmarks.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Thermography and Photoacoustic Techniques · Industrial Vision Systems and Defect Detection · Image Processing Techniques and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Physics and Electronics, Central South University, Changsha, Hunan, China
Physics Department, Capital Normal University, Beijing, China
School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin, China
Computer Vision and Systems Laboratory (CVSL), Department of Electrical and Computer Engineering, Laval University, Quebec, Canada
Centre for Composite Materials and Structures, Harbin Institute of Technology, Harbin, China
Funding
National Natural Science Foundation of China
References (40)
Show source-provided references
1. 10.1080/1058975070144866210.1080/10589750701448662 ↗
2. 10.1080/10589759.2019.159735610.1080/10589759.2019.1597356 ↗
3. 10.1080/10589759.2023.219728710.1080/10589759.2023.2197287 ↗
4. 10.1016/j.infrared.2021.10375410.1016/j.infrared.2021.103754 ↗
5. 10.1016/j.optlaseng.2021.10667910.1016/j.optlaseng.2021.106679 ↗
6. 10.1109/TII.2019.295471810.1109/TII.2019.2954718 ↗
7. 10.1016/S0035-3159(98)80048-210.1016/S0035-3159(98)80048-2 ↗
8. 10.1016/S0263-8223(02)00161-710.1016/S0263-8223(02)00161-7 ↗
9. 10.1088/0957-0233/24/12/12520510.1088/0957-0233/24/12/125205 ↗
10. 10.1109/tii.2018.288851910.1109/tii.2018.2888519 ↗
11. 10.1117/12.20485010.1117/12.204850 ↗
12. 10.1117/1.156696910.1117/1.1566969 ↗
13. 10.1063/1.473778410.1063/1.4737784 ↗
14. 10.1117/12.54085510.1117/12.540855 ↗
15. 10.1007/s10921-018-0535-810.1007/s10921-018-0535-8 ↗
16. 10.1016/j.ndteint.2009.05.00410.1016/j.ndteint.2009.05.004 ↗
17. 10.1016/j.infrared.2015.03.00310.1016/j.infrared.2015.03.003 ↗
18. 10.2478/mms-2013-004210.2478/mms-2013-0042 ↗
19. 10.1007/s11605-017-3518-710.1007/s11605-017-3518-7 ↗
20. 10.1016/J.NDTEINT.2005.04.00510.1016/J.NDTEINT.2005.04.005 ↗
21. 10.1016/S0963-8695(01)00041-X10.1016/S0963-8695(01)00041-X ↗
22. 10.1016/j.infrared.2019.10303210.1016/j.infrared.2019.103032 ↗
23. 10.1109/ACCESS.2020.301811610.1109/ACCESS.2020.3018116 ↗
24. 10.1007/s10921-022-00845-610.1007/s10921-022-00845-6 ↗
25. 10.3390/APP1019681910.3390/APP10196819 ↗
26. 10.1080/10589759.2023.220665510.1080/10589759.2023.2206655 ↗
27. 10.1109/CVPR.2015.729896510.1109/CVPR.2015.7298965 ↗
28. 10.1109/ICCV.2015.51010.1109/ICCV.2015.510 ↗
29. 10.1080/10589759.2022.211874710.1080/10589759.2022.2118747 ↗
30. 10.1080/10589759.2022.203073510.1080/10589759.2022.2030735 ↗
31. 10.1016/J.INFRARED.2021.10369410.1016/J.INFRARED.2021.103694 ↗
32. 10.1016/j.infrared.2019.10304810.1016/j.infrared.2019.103048 ↗
33. 10.1109/TII.2019.295226110.1109/TII.2019.2952261 ↗
34. 10.1109/TIM.2020.303496710.1109/TIM.2020.3034967 ↗
35. 10.1080/10589759.2023.223454810.1080/10589759.2023.2234548 ↗
36. 10.1016/j.eswa.2020.11324010.1016/j.eswa.2020.113240 ↗
37. 10.48550/arXiv.1506.0032710.48550/arXiv.1506.00327 ↗
38. 10.1063/1.147284810.1063/1.1472848 ↗
39. 10.1117/12.42103210.1117/12.421032 ↗
40. 10.1109/ACCESS.2020.302326210.1109/ACCESS.2020.3023262 ↗